Summary Membrane proteins that are critical for cell signaling and other biological processes reside in rafts, but rafts have been difficult to study because they have submicroscopic sizes, dynamic structures, and components that do not have fixed stoichiometries. The compositions of rafts - proteins and lipids - must be determined in order to understand which proteins interacting in cellular cascades come into proximity with each other as a result of their residence in a raft. A new experimental procedure now allows cellular rafts to be isolated at the physiological temperature of 37C, rather than the previously necessary temperature of 4C, enabling basic question in raft studies to be addressed. The compositions of many domains that exist within a cell membrane at mammalian body temperature (37C) are likely to be different from those of cell membranes that are kept on ice (4C), so many domains cannot be reliably determined at low temperature. The composition of rafts at biological temperature can now be determined, and membrane anchors of proteins favored in rafts can be compared to data derived at 4C. The relationship between types of proteins (GPI-anchored, transmembrane domain, prenylated) and amounts of cholesterol in a raft will be classified. Comparison of cholesterol levels in different types of rafts will uncover mechanisms that cause cholesterol to move between rafts. The displacement of one protein by another inside a raft, and the resulting effect on cellular processes, if any, will be assessed. The physical chemistry that controls the relationship between cholesterol and sphingomyelin content will be studied in a model raft system. The model system also allows the experimentally elusive question of how proteins contribute to raft formation to be approached in a concrete manner. For cellular studies, a new method will be employed that exploits the fact that the pressure needed to rupture a vesicle depends directly on the lipid composition of its membrane. Combining this method with the traditional approach of separation by membrane density will allow collection of a large range of domains. This will provide, for the first time, analysis of protein and lipid raft content without alteration caused by low temperature, detergents, and/or alkaline pH required by all previous raft isolation procedures. Isolating model bilayer domains containing a peptide and measuring compositions will determine the physical mechanisms that create these domains. This will yield experimentally testable hypotheses of mechanisms of biological domain formation.